Does advance booking and the internet age require a different timetable structure?

[Magdalia]

This is unfortunate, as one of the upsides of service changes in the last generation has been the development of better frequencies.

Of course, having provided Metro-like frequencies, the revenue team went the opposite way, with ticketing encouraging advance purchase for just one of the trains provided. The fools.

mods note - split from this thread

For me, this is the fundamental flaw in the way that the passenger railway operates. The timetable people have tried to make trains like buses, while the revenue people have tried to make trains like planes.

The passenger railway is stuck with a timetable based on the high frequency model, largely set in concrete at privatisation, but doubled down on afterwards, most notably on the WCML route modernisation. In the meantime the internet completely changed the way that many tickets are sold and how many travellers get information on what trains are running.

Coming from the Fens, the best example of this absurdity is the Liverpool-Norwich service, which runs hourly with 2 car trains, hogging capacity at some key pinch points, even though most of the long distance passengers have booked in advance.

For the railway to recover, it has to decide which way it is going to go on this, and then commit to it. There is an element of "horses for courses" here: for short distance high capacity routes the bus model is appropriate, for long distance book in advance routes the plane model is more appropriate, and there will be arguments at the margins.

But the franchise model is a huge obstacle to this because many franchises freeze the current route structure and timetable. In the past there were massive timetable upheavals in the 1960s following modernisation and in the 1980s following the introduction of Sprinters. We should have had a massive timetable upheaval in the 2010s, following the development of the internet. It didn't happen, because of franchising, and now the railway is paying the price.

 

[Austriantrain]

For me, this is the fundamental flaw in the way that the passenger railway operates. The timetable people have tried to make trains like buses, while the revenue people have tried to make trains like planes.

The passenger railway is stuck with a timetable based on the high frequency model, largely set in concrete at privatisation, but doubled down on afterwards, most notably on the WCML route modernisation. In the meantime the internet completely changed the way that many tickets are sold and how many travellers get information on what trains are running.

Coming from the Fens, the best example of this absurdity is the Liverpool-Norwich service, which runs hourly with 2 car trains, hogging capacity at some key pinch points, even though most of the long distance passengers have booked in advance.

For the railway to recover, it has to decide which way it is going to go on this, and then commit to it. There is an element of "horses for courses" here: for short distance high capacity routes the bus model is appropriate, for long distance book in advance routes the plane model is more appropriate, and there will be arguments at the margins.

But the franchise model is a huge obstacle to this because many franchises freeze the current route structure and timetable. In the past there were massive timetable upheavals in the 1960s following modernisation and in the 1980s following the introduction of Sprinters. We should have had a massive timetable upheaval in the 2010s, following the development of the internet. It didn't happen, because of franchising, and now the railway is paying the price.

I don‘t understand your point. How would an „internet-age timetable“ look like?

 

[Magdalia]

I don‘t understand your point. How would an „internet-age timetable“ look like?

Bizarrely, for long distance journeys it would look a bit like the days of steam. Any route where the tickets sold are primarily booked in advance should be running the longest trains possible, with lower frequencies.

Even on the busiest long distance routes, there's no need for anything more frequently than hourly if nearly everyone is boking in advance. If demand exceeds the capacity of one full length train, then run two or three in a flight. In the olden days they were called reliefs.

 

[JonathanH]

Even on the busiest long distance routes, there's no need for anything more frequently than hourly if nearly everyone is boking in advance. If demand exceeds the capacity of one full length train, then run two or three in a flight. In the olden days they were called reliefs.

In contrast, you can take the view that the 'internet age' and booking in advance against a timetable with a frequent service gives the passenger the choice to time their booking at the optimum time to suit their plans. It also allows the passenger to match up to onward connections.

Relief workings can't just be run in a flight, because the platform workings and siding space aren't available in the way that would facilitate them.

 

[The exile]

Bizarrely, for long distance journeys it would look a bit like the days of steam. Any route where the tickets sold are primarily booked in advance should be running the longest trains possible, with lower frequencies.

Even on the busiest long distance routes, there's no need for anything more frequently than hourly if nearly everyone is boking in advance. If demand exceeds the capacity of one full length train, then run two or three in a flight. In the olden days they were called reliefs.

That's fine as long as the long-distance services are precisely that - not (as is usually the case now) also an essential part of local connectivity as well. Local journeys may not bring a massive amount of revenue to the railway per journey (though they are of course more likely to be frequently repeated) - but they are more likely to be the ones taking cars off congested roads. There's also nothing to say that the "local journey" on a long distance train isn't also a crucial element of a long-distance journey that isn't covered by that through service - break connections, or reduce the frequency so much that the consequences of something going wrong are an inordinately long wait on a draughty platform, and lots of other long journeys suddenly disappear....

 

[Magdalia]

In contrast, you can take the view that the 'internet age' and booking in advance against a timetable with a frequent service gives the passenger the choice to time their booking at the optimum time to suit their plans.

Hourly services deliver that.

Relief workings can't just be run in a flight, because the platform workings and siding space aren't available in the way that would facilitate them.

It would be using the same capacity, but in a different way. A hypothetical example: instead of running Euston-Birmingham/Manchester/Liverpool every 20 minutes, there would be a Birmingham flight, then a Manchester flight, then a Liverpool flight, each with one, two or three trains according to demand.

That's fine as long as the long-distance services are precisely that - not (as is usually the case now) also an essential part of local connectivity as well.

You're right, and there's instances where capacity currently being taken by long distance services would be better used delivering local connectivity properly, Norwich-Liverpool being an example.

 

[JonathanH]

It would be using the same capacity, but in a different way. A hypothetical example: instead of running Euston-Birmingham/Manchester/Liverpool every 20 minutes, there would be a Birmingham flight, then a Manchester flight, then a Liverpool flight, each with one, two or three trains according to demand.

So, I think you are saying that London to Manchester should operate hourly but with gaps in the timetable to allow multiple trains to run possibly two and five minutes behind if demand requires. What would be the calling pattern of trains one, two and three? Presumably train one would run fast if two runs slower? Do local trains then run north from Stoke and Macclesfield in the rest of the hour to maintain their capacity to Manchester?

If London to Manchester runs once an hour, with relief trains, how frequently would Bournemouth to Manchester or London to Plymouth run?

 

[The Planner]

I thought the internet age had made us all want everything as soon as possible, Amazon Prime, Uber Eats, Deliveroo, Uber taxis. Therefore people want turn up and go?

 

[Starmill]

It would be using the same capacity, but in a different way. A hypothetical example: instead of running Euston-Birmingham/Manchester/Liverpool every 20 minutes, there would be a Birmingham flight, then a Manchester flight, then a Liverpool flight, each with one, two or three trains according to demand.

I'm afraid that it really wouldn't be the same capacity though.

How would you provide the Crewe / Wilmslow / Stoke-on-Trent / Macclesfield / Stockport - Manchester capacity if there's to be only one train in the flight that hour? Run several extras? Surely this would make every hour and day of the week different - a total logistical nightmare. If there are three within ten minutes, do you not then need a whole load of additional resources for the extra services in the rest of the hour?

 

[30907]

I'm afraid that it really wouldn't be the same capacity though.

How would you provide the Crewe / Wilmslow / Stoke-on-Trent / Macclesfield / Stockport - Manchester capacity if there's to be only one train in the flight that hour? Run several extras? Surely this would make every hour and day of the week different - a total logistical nightmare. If there are three within ten minutes, do you not then need a whole load of additional resources for the extra services in the rest of the hour?

You also need extra resources in terms of platforms at termini for the reliefs - eg 3 instead of 2 at Piccadilly.

In any case, the real inefficiencies are the example the OP quotes of Norwich-Nottingham which use up paths unnecessarily (#1) but should apparently still be hourly (#6).

 

[Magdalia]

How would you provide the Crewe / Wilmslow / Stoke-on-Trent / Macclesfield / Stockport - Manchester capacity if there's to be only one train in the flight that hour?

I would expect Manchester to need more than one train per hour most of the time on most days. The point is that they don't need to be at even intervals within each hour if most tickets are sold in advance.

You also need extra resources in terms of platforms at termini for the reliefs - eg 3 instead of 2 at Piccadilly.

I agree this is potentially an issue with more than 2 trains at country end termini such as Manchester and Liverpool.

In any case, the real inefficiencies are the example the OP quotes of Norwich-Nottingham which use up paths unnecessarily (#1) but should apparently still be hourly (#6).

I didn't say that everything should be hourly, only that, when most tickets are booked in advance, services do not need to be more frequent than that. Where the demand isn't sufficient for a full length train once per hour, then the service should be less frequent. For example, open access operators don't run hourly services.

If London to Manchester runs once an hour, with relief trains, how frequently would Bournemouth to Manchester or London to Plymouth run?

London-Plymouth would probably sustain an hourly service, Bournemouth-Manchester probably not.

 

[Starmill]

I would expect Manchester to need more than one train per hour most of the time on most days. The point is that they don't need to be at even intervals within each hour if most tickets are sold in advance.


I agree this is potentially an issue with more than 2 trains at country end termini such as Manchester and Liverpool.


I didn't say that everything should be hourly, only that, when most tickets are booked in advance, services do not need to be more frequent than that. Where the demand isn't sufficient for a full length train once per hour, then the service should be less frequent. For example, open access operators don't run hourly services.


London-Plymouth would probably sustain an hourly service, Bournemouth-Manchester probably not.

What you seem to be misunderstanding is that there's an over-riding requirement for London - Manchester trains to serve commuters into Manchester city centre on two routes, and London - West Midlands trains to serve commuters into Birmingham city centre from two directions. There is a smaller but significant requirement to provide capacity into Liverpool city centre and for short journeys between Crewe and Lancaster inclusive.

How can these markets be satisfied with your proposals?

 

[Magdalia]

I thought the internet age had made us all want everything as soon as possible, Amazon Prime, Uber Eats, Deliveroo, Uber taxis. Therefore people want turn up and go?

People still book restaurant tables and airport taxis in advance.

What you seem to be misunderstanding is that there's an over-riding requirement for London - Manchester trains to serve commuters into Manchester city centre, and London - West Midlands trains to serve commuters into Birmingham city centre.

There may be some usage like that, but it is definitely not an "over-riding requirement". The "over-riding requirement" for London-Manchester and London-West Midlands is the long distance journeys.

 

[Starmill]

There may be some usage like that, but it is definitely not an "over-riding requirement". The "over-riding requirement" for London-Manchester and London-West Midlands is the long distance journeys.

I'm afraid that's not the case. You've fallen into the very common trap which enthusiasts often do of just assuming that short traffic flows are dealt with separately to the long distance trains. This is so rarely the case for long distance services in this country. Even Lumo carry commuters the 15 minutes between Morpeth and Newcastle, and LNER's fastest trains, for now, still carry commuters 25 minutes between Darlington and Newcastle.

Only works on the scale of HS2 are able to achieve this kind of separation. And HS2 will now only relive services into London Euston, into Birmingham from multiple directions, and into Manchester from the south.

 

[The exile]

There may be some usage like that, but it is definitely not an "over-riding requirement". The "over-riding requirement" for London-Manchester and London-West Midlands is the long distance journeys.

That may be the largest single flow, but (without having the stats to hand) not necessarily the over-riding one (ie greater than all the other flows put together)

 

[Peter Lanky]

I would expect Manchester to need more than one train per hour most of the time on most days. The point is that they don't need to be at even intervals within each hour if most tickets are sold in advance.

I agree this is potentially an issue with more than 2 trains at country end termini such as Manchester and Liverpool.

As Eus-Man keeps being used as an example, there is another idea that has been much talked about in Lancashire but not it seems by the WCML. When a train arrives in Manchester, a significant number of passengers trek over to platform 14 to take a train to Bolton. This not only results in a very crowded platform 14, but when many of these passengers then alight at Bolton, the connecting train is almost empty. Yet with the recent electrifications, one train per hour could terminate in Bolton rather than Manchester to take pressure off platform 14 and make better use of the trains to Barrow, Windermere and Blackpool. The same in reverse.

I'm sure that the same could well apply in other parts of the country, where a large town is not on one of the main lines. It seems that large towns close to main cities are only considered as 'commute fodder' by the railways, with no thought that people may want to travel further afield without having to change trains too often.

 

[JonathanH]

When a train arrives in Manchester, a significant number of passengers trek over to platform 14 to take a train to Bolton. This not only results in a very crowded platform 14, but when many of these passengers then alight at Bolton, the connecting train is almost empty. Yet with the recent electrifications, one train per hour could terminate in Bolton rather than Manchester to take pressure off platform 14 and make better use of the trains to Barrow, Windermere and Blackpool. The same in reverse.

It would equally put a lot of pressure on platform 13 and 14 for all those passengers who want to travel to Manchester, rather than Bolton.

 

[Peter Lanky]

It would equally put a lot of pressure on platform 13 and 14 for all those passengers who want to travel to Manchester, rather than Bolton.

Not to the same extent, as they would alight the train and leave the platform immediately. When people are milling around waiting for 2 or 3 trains at 13/14 quite a crowd can congregate. The rail companies need to look for opportunities rather than look for objections.

 

[Magdalia]

I'm afraid that's not the case. You've fallen into the very common trap which enthusiasts often do of just assuming that short traffic flows are dealt with separately to the long distance trains. This is so rarely the case for long distance services in this country. Even Lumo carry commuters the 15 minutes between Morpeth and Newcastle, and LNER's fastest trains, for now, still carry commuters 25 minutes between Darlington and Newcastle.

I can assure you that I'm not looking at this as an enthusiast. I'm looking at it from the perspective of effective management of scarce resources. And I have 30 years of experience as a commuter.

I'm not assuming that short distance flows are dealt with separately to the long distance trains. But long distance Inter City trains are run primarily for long distance travellers. Lumo don't make their money from Morpeth-Newcastle passengers and LNER don't make their money from Darlington-Newcastle passengers. Both make their money from London-Newcastle and London-Edinburgh passengers.

Only works on the scale of HS2 are able to achieve this kind of separation. And HS2 will now only relive services into London Euston, into Birmingham from multiple directions, and into Manchester from the south.

Most routes to/from London do exactly that, without HS2 style separation. But release of capacity for commuter traffic in the West Midlands and Cheshire/Lancashire is an important part of the HS2 benefits.

 

[Bald Rick]

When a train arrives in Manchester, a significant number of passengers trek over to platform 14 to take a train to Bolton.

How many is ‘significant’? If it’s fewer than 150 people per train it’s not worth it.

I‘d be willing to bet that more people on Avanti’s Manchester services head for the Victoria Line at Euston than the Bolton train at Piccadilly.

 

[david1212]

Bizarrely, for long distance journeys it would look a bit like the days of steam. Any route where the tickets sold are primarily booked in advance should be running the longest trains possible, with lower frequencies.

Even on the busiest long distance routes, there's no need for anything more frequently than hourly if nearly everyone is booking in advance. If demand exceeds the capacity of one full length train, then run two or three in a flight. In the olden days they were called reliefs.

Where is the stock and staff for these reliefs going to come from? Core to the current cancellations / unreliability are issues over rest day working but if enough staff were prepared to work there are working hours regulations.

The stock of at least some reliefs was at best old and worn plus often not really suitable i.e. suburban used long distance. OK compartment stock has gone but putting all the technicalities of train/route compatibility and staff training aside just think of the reaction to e.g. sending an EMU designed for short suburban journeys from London up the WCML to Blackpool. 'Only' as far back as the 1970's if not later were Tyseley non-corridor and non-toilet DMU's run Birmingham Area - Blackpool and Pwllheli ( not the same train obviously ).

 

[Magdalia]

Where is the stock and staff for these reliefs going to come from? Core to the current cancellations / unreliability are issues over rest day working but if enough staff were prepared to work there are working hours regulations.

It is the existing rolling stock and traincrew redeployed, it does not require any new trains or drivers. It is just replacing trains at xx00, xx20 and xx40 with trains at xx00, xx03 and xx06, for example.

Where is the stock and staff for these reliefs going to come from? Core to the current cancellations / unreliability are issues over rest day working but if enough staff were prepared to work there are working hours regulations.

The stock of at least some reliefs was at best old and worn plus often not really suitable i.e. suburban used long distance. OK compartment stock has gone but putting all the technicalities of train/route compatibility and staff training aside just think of the reaction to e.g. sending an EMU designed for short suburban journeys from London up the WCML to Blackpool. 'Only' as far back as the 1970's if not later were Tyseley non-corridor and non-toilet DMU's run Birmingham Area - Blackpool and Pwllheli ( not the same train obviously ).

I'm definitely not suggesting those sorts of reliefs.

 

[WelshBluebird]

But long distance Inter City trains are run primarily for long distance travellers.

I'd disagree with that assumption. Some long distance intercity trains are. But many aren't. Many are run primarily for the short distance travellers on many different flows along that longer journey.

Lumo don't make their money from Morpeth-Newcastle passengers and LNER don't make their money from Darlington-Newcastle passengers.

Cross Country absolutely do make their money from the likes of Plymouth - Exeter, Exeter - Bristol, Bristol - Birmingham etc etc passengers.

GWR will make a significant amount of money on the Bristol - Bath - Swindon and Reading - London flows as another example.

 

[The Planner]

Are these relief trains timetabled, eating up capacity elsewhere if they don't run?

 

[david1212]

It is the existing rolling stock and traincrew redeployed, it does not require any new trains or drivers. It is just replacing trains at xx00, xx20 and xx40 with trains at xx00, xx03 and xx06, for example.

So instead of scheduling xx00, xx20 and xx40 you only schedule xx00 but if that will be beyond capacity you run xx03 and then if still demand xx06? If you only need the xx00 you save the direct running cost of the trains for the xx03 and xx06 but still have the capital cost.
How far ahead do you decide if the xx03 and xx06 will run so need the crew? How would you usefully use the crew if the xx03 and xx06 are not required? Or would you just have then on-call?

This all presumes only having a scheduled service at xx00 fits in with passenger needs. Once the one train an hour is not direct on average interchange times and hence the end-to-end journey times will increase. Even if direct it certainly goes against being able to get in your car and depart when either most convenient to you or to arrive no more than 20 minutes early rather than almost an hour early.

Take now XC south of Reading - Bournemouth only being every two hours instead of every hour even though if double sets the same capacity. When ever I have travelled it is the trains that are currently missing that fitted with my plans. Further if a cancellation or unrelated I miss the return then a wait of 2 hours rather than an hour. ( Yes other trains will get me to/from Reading with 1 or 2 changes but only 15 - 20 minutes ahead of the next XC ).

Or Chiltern xx00 from Birmingham then from Victoria xx00 to Brighton. Arrive Marylebone +1h 50min then tubes to Victoria arriving ~+2h 15min. Now 45 minutes wait instead of 15 if xx00 and xx30 ( close to actual times of fastest GX services ). OK everything can not leave the origin station xx00 but the average interchange wait will be longer.

 

[Starmill]

I can assure you that I'm not looking at this as an enthusiast. I'm looking at it from the perspective of effective management of scarce resources. And I have 30 years of experience as a commuter.

I'm not assuming that short distance flows are dealt with separately to the long distance trains. But long distance Inter City trains are run primarily for long distance travellers. Lumo don't make their money from Morpeth-Newcastle passengers and LNER don't make their money from Darlington-Newcastle passengers. Both make their money from London-Newcastle and London-Edinburgh passengers.



Most routes to/from London do exactly that, without HS2 style separation. But release of capacity for commuter traffic in the West Midlands and Cheshire/Lancashire is an important part of the HS2 benefits.

Certainly the money is all about the long distance trains. Nobody is disputing that. But the balance in terms of numbers is quite different.

If you're not carrying people making short journeys into Manchester and Birmingham centres on the various long-distance trains, there's simply no way those people can travel. They won't all fit on the three and four car stoppers, which are already all full with people travelling from the stations the long-distance trains don't stop at. That's what I meant about your idea for changing the service being totally impractical. Of course, with HS2 it will be different.

Now you can propose that you just don't want the sort of business of short journeys into Manchester and Birmingham, and that some smaller stations should close in order to support that if you like, and see how that goes down.

 

[whoosh]

How far ahead do you decide if the xx03 and xx06 will run so need the crew? How would you usefully use the crew if the xx03 and xx06 are not required? Or would you just have then on-call?

Indeed. Say the 13:00 Euston to Manchester has all tickets sold, so you now want to run the 13:03. There's only 5 tickets sold.
The 10:00 Manchester to Euston ran but wasn't full, so the 10:03 Manchester to Euston didn't run.
Do you run the 10:03 ECS from Manchester to Euston to make the 13:03 Euston to Manchester for the sake of 5 passengers or do you give them their money back and tell them to walk because they arent worth the cost?

Or,
Would you run a "relief" with low passenger numbers, but throw them all off (with a refund of course) and terminate short if there was suddenly a demand at an intermediate station in the other direction (perhaps a late finishing football match) in the style of 'pirate' buses in London in the 1920s?

 

[Bald Rick]

Or do you run a consistent service every day, of day three trains an hour, and manage demand with yield management?

 

[Dr Day]

The ability to re-book AP relatively easily on-the-day to adjacent trains with no fees would be a massive help. Of course, in some instances prices will have gone up and a relative differential may need to be paid, but if there is similar (low) demand on say the 1120 and the 1140 on a wet Wednesday in early February, and you have a ticket for the 1140 but rock up early (perhaps having been lucky with connections) at say 1110, then the ability to jump on the 1120 can make a huge difference to your journey - perhaps again being able to make connections at the other end getting home considerably earlier. Yes, some would say well if you want flexibility then you should buy a fully flexible ticket but with a significant difference in fare that isn't always affordable. And others would say just enjoy a pint in a nearby hostelry, but many people would rather simply get home.

Much of the benefit of investment in journey time and frequency improvements is lost if passengers feel they have to allow extra contingency to guarantee to make a specific trains, and are penalised for re-booking, even if there are more trains from them to chose from in the first place, and more seats overall.

ie still see the answer being on reform on the fares side, rather than the timetable.

 

[Magdalia]

So instead of scheduling xx00, xx20 and xx40 you only schedule xx00 but if that will be beyond capacity you run xx03 and then if still demand xx06? If you only need the xx00 you save the direct running cost of the trains for the xx03 and xx06 but still have the capital cost.

Yes, that's it.

How far ahead do you decide if the xx03 and xx06 will run so need the crew? How would you usefully use the crew if the xx03 and xx06 are not required? Or would you just have then on-call?

I'm not certain of the answer on that, but it would depend on balancing how soon it is possible to get good forecasts of demand and the planning horizon for fixing supply of rolling stock and traincrew resources. I wouldn't want to go as far as on-call working, but there might be scope for having fewer rest days at busy times and more rest days at quiet times.

Certainly the money is all about the long distance trains. Nobody is disputing that.

And where the railway is now, it is the money that talks.

Indeed. Say the 13:00 Euston to Manchester has all tickets sold, so you now want to run the 13:03. There's only 5 tickets sold.
The 10:00 Manchester to Euston ran but wasn't full, so the 10:03 Manchester to Euston didn't run.
Do you run the 10:03 ECS from Manchester to Euston to make the 13:03 Euston to Manchester for the sake of 5 passengers or do you give them their money back and tell them to walk because they arent worth the cost?

Yield management wouldn't go away, if anything it would become more important. Yield management would still be needed to avoid running trains with 5 passengers, and it would still be needed to ensure that rolling stock and traincrew were not idle. But it has to be a step forward to be able to manage supply and demand not just demand.

if there was suddenly a demand at an intermediate station in the other direction (perhaps a late finishing football match)

I'm glad someone has mentioned football. Flexibility of supply could be a significant benefit on football matchdays. For example, if Liverpool are kicking off at 1230 and Manchester United at 1500 then early morning capacity from London can be weighted towards Liverpool and later morning capacity towards Manchester. Similar considerations could apply to other big sporting and cultural events.

There's a huge amount of data on ticketing and travelling now, and the computer power to process it. That opens up potential for the railway to manage demand and supply in ways that weren't possible 20-30 years ago, driving revenue up and costs down. Yield management has been doing that on the demand side for a while now. The supply side is more of a challenge, but every crisis is an opportunity, and the railway is in crisis now.

This has strayed from my original point, which was that, when most tickets are advance booked, 3 trains per hour do not need be every 20 minutes.